Détail de l'auteur
Auteur Yuan Tian |
Documents disponibles écrits par cet auteur (5)
Ajouter le résultat dans votre panier Affiner la recherche Interroger des sources externes
Automatic spatial–spectral feature selection for hyperspectral image via discriminative sparse multimodal learning / Qian Zhang in IEEE Transactions on geoscience and remote sensing, vol 53 n° 1 (January 2015)
[article]
Titre : Automatic spatial–spectral feature selection for hyperspectral image via discriminative sparse multimodal learning Type de document : Article/Communication Auteurs : Qian Zhang, Auteur ; Yuan Tian, Auteur ; Yuan Yang, Auteur ; Chunhong Pan, Auteur Année de publication : 2015 Article en page(s) : pp 261 - 279 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] analyse discriminante
[Termes IGN] apprentissage (cognition)
[Termes IGN] apprentissage dirigé
[Termes IGN] classification spectrale
[Termes IGN] image hyperspectrale
[Termes IGN] matrice
[Termes IGN] méthode des moindres carrés
[Termes IGN] programmation non linéaireRésumé : (Auteur) Spectral-spatial feature combination for hyperspectral image analysis has become an important research topic in hyperspectral remote sensing applications. A simple and straightforward way to integrate spectral-spatial features is to concatenate heterogeneous features into a long vector. Then, the dimensionality reduction techniques, i.e., feature selection, are applied before subsequent utilizations. However, such representation can introduce redundancy and noise. Moreover, traditional single-feature selection methods treat different features equally and ignore their complementary properties. As a result, the performance of subsequent tasks, i.e., classification, would drop down. In this paper, we propose a novel approach to integrate the spectral-spatial features based on the concatenating strategy, termed discriminative sparse multimodal learning for feature selection (DSML-FS). In the proposed method, joint structured sparsity regularizations are used to exploit the intrinsic data structure and relationships among different features. Discriminative least squares regression is applied to enlarge the distance between classes. Therefore, the weight matrix incorporating the information of feature wise and individual properties is automatically learned for spectral-spatial feature selection. We develop an alternative iterative algorithm to solve the nonlinear optimization problem in DSML-FS with global convergence. We systematically evaluate the proposed algorithm on three available hyperspectral data sets, and the encouraging experimental results demonstrate the effectiveness of DSML-FS. Numéro de notice : A2015-032 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2014.2321277 En ligne : https://doi.org/10.1109/TGRS.2014.2321277 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=75114
in IEEE Transactions on geoscience and remote sensing > vol 53 n° 1 (January 2015) . - pp 261 - 279[article]Réservation
Réserver ce documentExemplaires(1)
Code-barres Cote Support Localisation Section Disponibilité 065-2015011 RAB Revue Centre de documentation En réserve L003 Disponible Evaluation of a satellite-based global flood monitoring system / K. Yilmaz in International Journal of Remote Sensing IJRS, vol 31 n° 14 (July 2010)
[article]
Titre : Evaluation of a satellite-based global flood monitoring system Type de document : Article/Communication Auteurs : K. Yilmaz, Auteur ; R. Adler, Auteur ; Yuan Tian, Auteur ; et al., Auteur Année de publication : 2010 Article en page(s) : pp 3763 - 3782 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection
[Termes IGN] erreur systématique
[Termes IGN] image satellite
[Termes IGN] image TRMM-MI
[Termes IGN] inondation
[Termes IGN] modèle hydrographique
[Termes IGN] prévention des risques
[Termes IGN] risque naturel
[Termes IGN] ruissellement
[Termes IGN] surveillance hydrologiqueRésumé : (Auteur) This study provides an initial evaluation of a global flood monitoring system (GFMS) using satellite-based precipitation and readily available geospatial datasets. The GFMS developed by our group uses a relatively simple hydrologic model, based on the run-off curve number method, to transform precipitation into run-off. A grid-to-grid routing scheme moves run-off downstream. Precipitation estimates are from the TRMM Multi-satellite Precipitation Analysis (TMPA). We first evaluated the TMPA algorithm using a radar/gauge merged precipitation product (Stage IV) over south-east USA. This analysis indicated that the spatial scale (and hence the basin size) as well as regional and seasonal considerations are important in using the TMPA to drive hydrologic models. GFMS-based run-off simulations were evaluated using observed streamflow data at the outlet of two US basins and also using a global flood archive. Basin-scale analysis showed that the GFMS was able to simulate the onset of flood events produced by heavy precipitation; however, the simulation performance deteriorated in the later stages. This result points out the need for an improved routing component. Global-scale analysis indicated that the GFMS is able to detect 38% of the observed floods; however, it suffers from region-dependent bias. Numéro de notice : A2010-370 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/01431161.2010.483489 En ligne : https://doi.org/10.1080/01431161.2010.483489 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=30564
in International Journal of Remote Sensing IJRS > vol 31 n° 14 (July 2010) . - pp 3763 - 3782[article]Exemplaires(1)
Code-barres Cote Support Localisation Section Disponibilité 080-2010091 RAB Revue Centre de documentation En réserve L003 Exclu du prêt Knowledge-based building reconstruction from terrestrial video sequences / Yuan Tian in ISPRS Journal of photogrammetry and remote sensing, vol 65 n° 4 (July - August 2010)
[article]
Titre : Knowledge-based building reconstruction from terrestrial video sequences Type de document : Article/Communication Auteurs : Yuan Tian, Auteur ; Markus Gerke, Auteur ; M. George Vosselman, Auteur ; Q. Zhu, Auteur Année de publication : 2010 Article en page(s) : pp 395 - 408 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] base de connaissances
[Termes IGN] image vidéo
[Termes IGN] prise de vue terrestre
[Termes IGN] reconstruction 3D du bâti
[Termes IGN] séquence d'imagesRésumé : (Auteur) The paper presents an automatic method for the reconstruction of building models from video image sequences. These videos may be recorded using a hand-held camera or a camera mounted on a moving car. Such terrestrial video sequences are economic and flexible. Presenting buildings as geometric models–rather than for instance a representation from a simple meshing of 3D points–enables one to perform a wide range of analyses. However, sparse 3D points and 3D edges do not contain topological relations. Therefore, integrating building structure knowledge into the reconstruction steps plays an important role in our method. First, some rules are applied to reasonably group the extracted features. Then, a suitable outline and normal direction are specified for each surface patch. Based on these surface patches, a hybrid model- and data-driven method is used to recover a building model from both the extracted surface patches and hypothesized parts. Using the building structure knowledge leads to a simple and fast reconstruction method, and also enables one to obtain the main structures of buildings. The results show that this method correctly sets up topological relationships between generated surface patches and also obtains reasonable structure models in occluded areas. Therefore, the reconstructed models satisfy requirements for both visualization and analysis. Numéro de notice : A2010-299 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2010.05.001 En ligne : https://doi.org/10.1016/j.isprsjprs.2010.05.001 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=30493
in ISPRS Journal of photogrammetry and remote sensing > vol 65 n° 4 (July - August 2010) . - pp 395 - 408[article]Réservation
Réserver ce documentExemplaires(1)
Code-barres Cote Support Localisation Section Disponibilité 081-2010041 SL Revue Centre de documentation Revues en salle Disponible An integrated method for satellite image interpolation / Wei Shi in International Journal of Remote Sensing IJRS, vol 28 n°5-6 (March 2007)
[article]
Titre : An integrated method for satellite image interpolation Type de document : Article/Communication Auteurs : Wei Shi, Auteur ; Yuan Tian, Auteur Année de publication : 2007 Article en page(s) : pp 1355 - 1371 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] image à basse résolution
[Termes IGN] interpolation bicubique
[Termes IGN] interpolation bilinéaireRésumé : (Auteur) Many image interpolation methods have been developed to enhance the visibility of either remotely sensed or other images, such as the bilinear method and bi-cubic method As a further development of the existing approaches, an integrated interpolation method is proposed in this study, which is an integration of the bilinear and the bi-cubic interpolation methods. First, the implemented procedure of the integrated interpolation method is described. This covers (a) a low-resolution image that is interpolated by using the bilinear and the bi-cubic interpolators respectively; (b) the hybrid parameters and 1 - endowed on the bilinear and the bi-cubic interpolation results, respectively; (c) the interpolated image that is computed according to the weighted sum of both bilinear and bi-cubic interpolation results. Second, a further discussion is given on the method and the relation between hybrid parameter and details of an image (entropy) provided from the theoretical point of view. The result demonstrates that the hybrid parameter directly affects the details of an interpolated image. Third, the effectiveness of this integrated method is verified, based on experimental studies. Here, a method for the construction of images with different entropies is developed for construct simulated images. The integrated method possesses advantages of both the bilinear method for modelling low-frequency components and the bi-cubic method of high-frequency components simultaneously. The experimental study demonstrated the effectiveness of the proposed integrated model. Copyright Taylor & Francis Numéro de notice : A2007-098 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/01431160600851876 En ligne : https://doi.org/10.1080/01431160600851876 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=28463
in International Journal of Remote Sensing IJRS > vol 28 n°5-6 (March 2007) . - pp 1355 - 1371[article]Réservation
Réserver ce documentExemplaires(1)
Code-barres Cote Support Localisation Section Disponibilité 080-07031 RAB Revue Centre de documentation En réserve L003 Disponible A hybrid interpolation method for the refinement of a regular grid digital elevation model / W.Z. Shi in International journal of geographical information science IJGIS, vol 20 n° 1 (january 2006)
[article]
Titre : A hybrid interpolation method for the refinement of a regular grid digital elevation model Type de document : Article/Communication Auteurs : W.Z. Shi, Auteur ; Yuan Tian, Auteur Année de publication : 2006 Article en page(s) : pp 53 - 67 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications photogrammétriques
[Termes IGN] grille régulière
[Termes IGN] interpolation linéaire
[Termes IGN] interpolation non-linéaire
[Termes IGN] matrice
[Termes IGN] modèle numérique de surface
[Termes IGN] système d'information géographiqueRésumé : (Auteur) A digital elevation model (DEM), which is used to represent a terrain surface. is normally constructed by applying an interpolation method on given sample elevation points. Interpolation methods can be classified into two classes: linear methods, which have a low time cost and are suitable for terrains where there is little change in elevation, and non-linear methods, which normally consume comparatively more time and are more suitable for terrains where there are frequent changes in elevation. A hybrid interpolation method, which involves both a linear method and a non-linear method of interpolation, is proposed in this paper. The proposed method aims to integrate the advantages of both linear and non-linear interpolation methods for the refinement of regular grid DEM. Here, the bilinear is identified as the linear method, and the bi-cubic is taken to be the non-linear interpolation method. The hybrid method is an integration of a linear model and non-linear interpolation model with a parameter that defines the weights for each of the models. The parameter is dependent on the complexity of the terrain, for which a DEM is to be interpolated. The experimental results in this study demonstrate that the hybrid method is effective for interpolating DEMs for various types of terrain. Numéro de notice : A2006-025 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1080/13658810500286943 En ligne : https://doi.org/10.1080/13658810500286943 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=27752
in International journal of geographical information science IJGIS > vol 20 n° 1 (january 2006) . - pp 53 - 67[article]Réservation
Réserver ce documentExemplaires(2)
Code-barres Cote Support Localisation Section Disponibilité 079-06011 RAB Revue Centre de documentation En réserve L003 Disponible 079-06012 RAB Revue Centre de documentation En réserve L003 Disponible